Awareness
LSTN Editorial Team · Editorially overseen by Dan McCoy
Most people who struggle to follow conversations hear reasonably well. The problem isn't detecting sound, it's processing speech. These two things are measured differently, affected by different parts of the auditory system, and require different solutions.
The ear's job is to detect sound. The brain's job is to decode speech. These two systems can fail independently, and often do.
Hearing sensitivity is measured in decibels: how loud a sound needs to be before you can detect it. But detecting a sound is not the same as understanding it. Speech clarity depends on frequency-specific sensitivity and on how well the auditory cortex processes what arrives. Both can be intact for one and impaired for the other.
This is why so many people say 'I can hear, but I can't understand.' They're correct. Both things can be true at the same time. An audiogram can show relatively normal thresholds while speech discrimination scores are significantly reduced.
Most hearing loss starts in the high frequencies: 2,000 Hz and above. This is exactly where consonants live: the /s/, /f/, /sh/, /th/ sounds that carry the semantic meaning in English words.
When sensitivity decreases in that range, vowels still come through clearly. They're louder and lower-pitched. But consonants fade. 'Ship' sounds like 'hip.' 'Coffee' sounds like 'copy.' The volume isn't wrong; the word is incomplete.
This explains the classic presentation: someone hears their spouse talking but keeps mishearing individual words. They're not imagining it and they're not being inattentive. The speech is physically incomplete by the time it reaches their auditory cortex.
In a quiet room, the brain can fill in gaps using context. If you miss the /s/ in 'ship,' the sentence usually makes the word clear. But in a noisy restaurant, two things happen simultaneously: the signal-to-noise ratio drops (speech becomes harder to separate from background), and the cognitive load increases, leaving less processing capacity available for decoding.
People with hearing loss also lose the ability to selectively attend to one voice among many. Normal-hearing listeners do this automatically. The ability to track a single voice in a crowd is called the 'cocktail party effect.' Many people with any degree of high-frequency loss cannot.
The result is a person who functions fine in a one-on-one conversation in a quiet room, but feels completely lost at a dinner party or a large meeting. This is not a personality issue or social anxiety. It's a specific acoustic phenomenon.
In the clinics I managed, the complaint was almost always the same: 'I'm fine at home, but I can't follow anything at a restaurant.' The audiologists on my team would note that this pattern was the hallmark of early high-frequency loss: quiet environments still manageable, noisy ones not. The audiogram often showed mild loss. The daily experience was already significant.
A standard audiogram measures hearing thresholds: how soft a tone needs to be before you can detect it at each frequency. It does not directly measure how well the auditory system decodes speech.
Speech discrimination testing (sometimes called word recognition testing) measures that. An audiologist presents recorded words at a comfortable volume and asks you to repeat them. Scores below 80% are clinically significant. Some people with relatively normal audiograms score poorly on word recognition.
There's also auditory processing disorder (APD), where hearing thresholds are normal but the brain struggles to interpret speech correctly. This is distinct from hearing loss and requires a different evaluation, typically by an audiologist trained in central auditory processing assessment.
Common Questions